检索结果-内蒙古大学图书馆

Empirical studies on the properties of linear regions in deep neural networks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhang, Xiao Wu, Dongrui Ministry of Education Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

A deep neural network (DNN) with piecewise linear activations can partition the input space into numerous small linear regions, where different linear functions are fitted. It is believed that the number of these regions represents the expressivity of the DNN. This paper provides a novel and meticulous perspective to look into DNNs: Instead of just counting the number of the linear regions, we study their local properties, such as the inspheres, the directions of the corresponding hyperplanes, the decision boundaries, and the relevance of the surrounding regions. We empirically observed that different optimization techniques lead to completely different linear regions, even though they result in similar classification accuracies. We hope our study can inspire the design of novel optimization techniques, and help discover and analyze the behaviors of DNNs. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

A fusion algorithm of remote sensing image based on discrete wavelet packet

学校读者我要写书评

暂无评论

A fusion algorithm of remote sensing image based on discrete...

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Hai-Hui Wang Jia-Xiong Peng Wei Wu State Key Laboratory for Image Processing and Intelligent Control Institute for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)0780378652

Image fusion refers to the techniques that integrate complementary information from multiple image sensor data such that the new images are more suitable for the purpose of human visual perception and the compute processing tasks. In this paper, a new image fusion algorithm based on wavelet packet transform to fuse multisensor images is presented. Discrete wavelet transform (DWT) can offer a more precise way for image analysis, than other multi-resolution analysis. It decomposes an image into low frequency band and high frequency band in different level, and it can also be reconstructed gradually in different level. But this method only decomposes low frequency band in a higher scale, so that it omits some useful details of the images. In this paper, we present a new image fusion algorithm. In the algorithm, we use discrete wavelet packet transform (DWPT) to decompose and reconstruct the images. When images are merged in wavelet packet space, different frequency ranges are processed differently. It can merge information from original images adequately and improve abilities of information analysis and feature extraction. This image fusion is performed at the pixel level. In this fusion algorithm, a feature-based fusion rule is used to combine original subimages and to form a pyramid for the fused image. Through merging remote sensing images from multi-sensor to a same object by applying method of wavelet packet analysis, we have obtained a fused picture. In this paper, mutual information is employed as a means of objective assessing image fusion performance. The experiment results show that this fusion algorithm, based on wavelet packet transform, is an effective approach in image fusion area.

关键词： Remote sensing Discrete wavelet transforms Wavelet packets Image fusion Frequency Image analysis Wavelet analysis Image reconstruction Image sensors Humans

Hybrid Relation Guided Set Matching for Few-shot Action Recognition

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Xiang Zhang, Shiwei Qing, Zhiwu Tang, Mingqian Zuo, Zhengrong Gao, Changxin Jin, Rong Sang, Nong Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Alibaba Group China

Current few-shot action recognition methods reach impressive performance by learning discriminative features for each video via episodic training and designing various temporal alignment strategies. Nevertheless, they are limited in that (a) learning individual features without considering the entire task may lose the most relevant information in the current episode, and (b) these alignment strategies may fail in misaligned instances. To overcome the two limitations, we propose a novel Hybrid Relation guided Set Matching (HyRSM) approach that incorporates two key components: hybrid relation module and set matching metric. The purpose of the hybrid relation module is to learn task-specific embeddings by fully exploiting associated relations within and cross videos in an episode. Built upon the task-specific features, we reformulate distance measure between query and support videos as a set matching problem and further design a bidirectional Mean Hausdorff Metric to improve the resilience to misaligned instances. By this means, the proposed HyRSM can be highly informative and flexible to predict query categories under the few-shot settings. We evaluate HyRSM on six challenging benchmarks, and the experimental results show its superiority over the state-of-the-art methods by a convincing margin. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

Integer Haar wavelet for remote sensing image compression

学校读者我要写书评

暂无评论

Integer Haar wavelet for remote sensing image compression

International Conference on Signal processing Proceedings (ICSP)

作者： Wang Zhenhua Xu Hongbo Tian Yan Tian Jinwen Liu Jian Institute for Pattern Recognition and Artificial Intelligence State Key Laboratory for Image Processing and Intelligent Control Huazhong University of science and Technology Wuhan China

ISBN: (纸本)0780374886

With the development of remote sensing technique, onboard data compression has become an urgent need and a lot of study has been directed toward the development of efficient techniques. In this paper, the construction approach of integer Haar is discussed briefly, and a simple image compression scheme based on the integer Haar wavelet transform and block DPCM is proposed. The scheme can be easily designed for data processing in real-time systems of remote sensing with parallel algorithms. Simulation experimental results demonstrate that the proposed approach is a efficient image compression method.

关键词： Image coding Remote sensing Wavelet transforms Frequency Data compression Image reconstruction Pattern recognition Artificial intelligence Image processing Intelligent control

Representative graph neural network

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Yu, Changqian Gao, Changxin Sang, Nong Yu, Changqian Liu, Yifan Shen, Chunhua Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science & Technology China University of Adelaide Australia

Non-local operation is widely explored to model the long-range dependencies. However, the redundant computation in this operation leads to a prohibitive complexity. In this paper, we present a Representative Graph (RepGraph) layer to dynamically sample a few representative features, which dramatically reduces redundancy. Instead of propagating the messages from all positions, our RepGraph layer computes the response of one node merely with a few representative nodes. The locations of representative nodes come from a learned spatial offset matrix. The RepGraph layer is flexible to integrate into many visual architectures and combine with other operations. With the application of semantic segmentation, without any bells and whistles, our RepGraph network can compete or perform favourably against the state-of-the-art methods on three challenging benchmarks: ADE20K, Cityscapes, and PASCAL-Context datasets. In the task of object detection, our RepGraph layer can also improve the performance on the COCO dataset compared to the non-local operation. Code is available at https://***/RepGraph. Copyright © 2020, The Authors. All rights reserved.

关键词： Semantic Segmentation

Knowledge-Data Fusion Based Source-Free Semi-Supervised Domain Adaptation for Seizure Subtype Classification

学校读者我要写书评

暂无评论

Knowledge-Data Fusion Based Source-Free Semi-Supervised Doma...

IEEE International Conference on Systems, Man and Cybernetics

作者： Ruimin Peng Jiayu An Dongrui Wu Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9781665410205

ISBN: (纸本)9781665410212

Electroencephalogram (EEG)-based seizure sub-type classification enhances clinical diagnosis efficiency. Source-free semi-supervised domain adaptation (SF-SSDA), which transfers a pre-trained model to a new dataset with no source data and limited labeled target data, can be used for privacy-preserving seizure subtype classification. This paper considers two challenges in SF-SSDA for EEG-based seizure subtype classification: 1) How to effectively fuse both raw EEG data and expert knowledge in classifier design? 2) How to align the source and target domain distributions for SF-SSDA? We propose a Knowledge-Data Fusion based SF-SSDA approach, KDF-MutuaISHOT, for EEG-based seizure subtype classification. In source model training, KDF uses Jensen-Shannon Diver-gence to facilitate mutual learning between a feature-driven Decision Tree-based model and a data-driven Transformer-based model. To adapt KDF to a new target dataset, an SF-SSDA algorithm, MutualSHOT, is developed, which features a consistency-based pseudo-label selection strategy. Experiments on the public TUSZ and CHSZ datasets demonstrated that KDF-MutualSHOT outperformed other supervised and source-free domain adaptation approaches in cross-subject seizure subtype classification.

关键词： Training Adaptation models Data privacy Fuses Brain modeling Transformers Data models Electroencephalography Decision trees Cybernetics

Super Twisting control of Passive Gait Training Exoskeleton Driven by Pneumatic Muscles

学校读者我要写书评

暂无评论

Super Twisting Control of Passive Gait Training Exoskeleton ...

International Symposium on Micromechatronics and Human Science (MHS)

作者： Mengshi Zhang Jian Huang Yu Cao Hai-Tao Zhang Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9781728126753

ISBN: (纸本)9781728126760

Pneumatic Muscles (PMs) - driven exoskeleton has a promising prospect in the field of rehabilitation and assistance, because of the PMs intrinsic features of compliance and high force-to-weight ratio. However, the precise control of PMs-driven exoskeleton still remains a challenging problem due to the hysteresis, time-varying parameters of PMs and complexity of the mechanism. To solve this problem, this paper presents a super twisting controller (STC), essentially a second-order sliding mode control, to realize the human gait tracking control of lower limb exoskeleton. As a result, the disturbances and uncertainties of the system can be handled, and the chattering caused by the traditional sliding mode control (SMC) could be eliminated. The stability of the closed-loop system is ensured according to the Lyapunov theorem. In addition, we conduct experiments in real exoskeleton system and the results illustrate the validity of the super twisting algorithm..

关键词： Uncertainty Heuristic algorithms Exoskeletons Muscles Stability analysis Closed loop systems Sliding mode control

Different Set Domain Adaptation for Brain-Computer Interfaces: A Label Alignment Approach

学校读者我要写书评

暂无评论

arXiv 2019年

作者： He, He Wu, Dongrui Ministry of Education Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

A brain-computer interface (BCI) system usually needs a long calibration session for each new subject/task to adjust its parameters, which impedes its transition from the laboratory to real-world applications. Domain adaptation, which leverages labeled data from auxiliary subjects/tasks (source domains), has demonstrated its effectiveness in reducing such calibration effort. Currently, most domain adaptation approaches require the source domains to have the same feature space and label space as the target domain, which limits their applications, as the auxiliary data may have different feature spaces and/or different label spaces. This paper considers different set domain adaptation for BCIs, i.e., the source and target domains have different label spaces. We introduce a practical setting of different label sets for BCIs, and propose a novel label alignment (LA) approach to align the source label space with the target label space. It has three desirable properties: 1) LA only needs as few as one labeled sample from each class of the target subject;2) LA can be used as a preprocessing step before different feature extraction and classification algorithms;and, 3) LA can be integrated with other domain adaptation approaches to achieve even better performance. Experiments on two motor imagery datasets demonstrated the effectiveness of LA. Copyright © 2019, The Authors. All rights reserved.

关键词： Brain computer interface

Demand-Side Frequency control in Power Network Based on Multi-Agent Deep Reinforcement Learning

学校读者我要写书评

暂无评论

Demand-Side Frequency Control in Power Network Based on Mult...

International Conference on Industrial Artificial intelligence (IAI)

作者： Xianggang Liu Zhi-Wei Liu Ming Chi Yifan Zhang Key Laboratory of Image Processing and Intelligent Control Ministry of Education College of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

The integration of renewable energy sources into the power grid has led to new challenges in maintaining the stability of the system frequency. This paper proposes a novel approach to address the Optimal Demand Side Frequency control (ODFC) problem using Multi-Agent Deep Deterministic Policy Gradient (MADDPG) method. The proposed method models the ODFC problem as a Markov game, with centralized training based on multi-agent cooperative self-learning and associative storage service. In the decentralized execution stage, each agent independently outputs control actions to the controlled plant using local observations. Numerical simulations show that the proposed method effectively addresses the ODFC problem with superior performance compared to traditional methods.

关键词：